Separations of lipids by silver

ion chromatography
L. J. MORRIS
The Biosynthesis Unit, Unilever Research Laboratory, Colworth House, Shambrook, Bedford, England
ABSTRACT The possibility of separating lipid materials on
the basis of the number, type, and position of the unsaturated
centers they contain, by virtue of the complexing of these un-
saturated bonds with silver ions, provides a relatively recent
but now very important addition to the range of separatory
methods available to lipid chemists and biochemists. In this
review, the nature of the complexing of silver ions with olefins
is considered briefly and the history of the development of
separation methods based on argentation is traced. Someprac-
tical considerations of argentation chromatography are dis-
cussed and separations of fatty acids and aldehydes, substi-
tuted fatty acids, neutral lipids, polar lipids, and sterols and
other terpenoid compounds, by argentation methods alone
and in conjunction with other separation techniques, are then
reviewed. Some conclusions are finally presented as to the
present and potential utility of argentation methods in studies
of the occurrence, metabolism, and function of lipids.
KEY WORDS lipid separations . argentation *
silver-olefin complexes . thin-layer . column
. paper . chromatography . fattyacids 9 neutral
lipids . polar lipids sterols . terpenoid
compounds
I T Is ONLY some 4 or 5 years since separations of lipo-
philic materials according to unsaturation by chromato-
graphic or countercurrent distribution methods involving
silver ions were first reported. Already separation meth-
ods based on complexing of unsaturated centers with
silver ions are in use in most laboratories engaged in
lipid research and more than 150 publications have ap-
Abbreviations: TLC, thin-layer chromatography ; GLC, gas-
liquid chromatography.
peared describing separations by such methods or their
utilization in lipid research. I t is now probably true to
say that argentation chromatography is third only to
gas-liquid chromatography (GLC) and normal
thin-layer chromatography in importance as a separatory
tool for natural lipid materials. That argentation chro-
matographic methods have gained such wide acceptance
and popularity so quickly indicates that there was a
widely felt need for a simple method of separating lipids
according to degree of unsaturation. The basic simplicity
of the method and the ready availability of such equip-
ment as is required have also encouraged its rapid ac-
ceptance.
The applications of separation methods involving
silver ions for the separation of lipids have been specifi-
cally reviewed a number of times (1-5) and have been
summarized in several other general reviews of lipid
separations (e.g. 6-13). However, few of these reviews
were at all comprehensive and since new procedures and
applications have been appearing almost monthly most
of them are now somewhat out of date. I n the present
review, the nature of the complexes between silver ions
and unsaturated centers is first considered and the de-
velopment of the various separation methods depending
on the formation of such complexes is briefly traced.
After consideration of some practical aspects of the var-
ious methods, the bulk of the review describes applica-
tions of these methods to separations within the various
categories of lipids. Separation of sterols and other lipo-
philic materials is also considered. While no guarantee
is made that all publications which have described
separations by argentation chromatography are in-
cluded, it is hoped that most of the important applica-
tions to date have been considered and critically assessed.
JOURNAL OF LIPID RESEARCH VOLUME 7, 1966 717

b
y

g
u
e
s
t
,

o
n

S
e
p
t
e
m
b
e
r

1
6
,

2
0
1
4
w
w
w
.
j
l
r
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

NATURE OF THE SI LVER
I ON COMPLEXES
It has been known for many years that there are weak
interactions between silver ions and certain compounds
containing ethylenic or acetylenic bonds. This phe-
nomenon was first quantitatively studied in 1938 by
Winstein and Lucas (1 4). They determined equilibrium
constants for the reaction of silver ions with a number of
acyclic and alicyclic olefinic compounds by measuring
the partitioning of the olefin between an aqueous silver
nitrate phase and carbon tetrachloride. Many studies
of the equilibrium constants of these reactions have been
carried out since, using either the original partition
method, or a method in which the solubility of the pure
unsaturated compound in aqueous solutions of silver
salts is determined, or, more recently, chromatographic
methods, notably GLC on solutions of silver nitrate in
suitable glycols as stationary phase (15,16). These basic
chemical studies were done principally by three.,groups :
Lucas and his coworkers (17) dealt chiefly with aliphatic
olefins and alkynes, Andrews and Keefer (18) were
concerned with aromatic compounds, and Traynham
and coworkers (1 9) have studied alicyclic olefins. Many
other groups, of course, have contributed to current
knowledge of complexing between centers of unsatura-
tion and silver and other transition nietal ions, and the
effects of substitution, stereochemistry, ring size, etc. on
the stability of such complexes. Many complexes have
been isolated in crystalline and relatively stable form,
particularly those between silver nitrate, silver per-
chlorate, and silver fluoroborate and cyclic olefins. This
review, however, is primarily concerned with chro-
matographic separations of lipids that depend on silver
ion complexing and although much can be learned from
these chemical studies of relatively simple compounds,
particularly by those concerned with separations and
structure elucidations of sterols and other alicyclic olefins,
more detailed consideration here of their findings and
implications seems inappropriate. More detailed in-
formation may be obtained from the original publica-
tions and from a number of reviews (e.g. 20-22).
The nature of the bonding in complexes of unsaturated
compounds with silver (and other transition metals with
nearly filled d-orbitals) has been a matter for some dis-
cussion. I t was originally suggested (14) that the [ole-
fin Ag]+species could be represented as a resonance
hybrid of three forms, later increased to four forms
(17a) :
This represents a simple r-complex in which only de-
formation of the r-orbitals of the olefin is involved.
An alternative picture, which now seems to be generally
accepted, was presented by Dewar (23) :
(11)
The bonding is considered to involve a u-bond formed
by overlap of the filled r-orbital of the olefin with (in the
case of silver) the free s-orbital (I) and a r-bond formed
by overlap of the vacant antibonding ?r-orbitals of the
olefin with filled d-orbitals of the silver (11). The bond-
ing in the complex will be affected by the availability of
electrons in the filled orbitals and the ease of overlap of
these orbitals, which is determined by steric factors.
On the basis of the Dewar concept, complexing with sil-
ver should not cause great change in the double bond.
That the double bond does remain almost intact in such
complexes is indicated by Raman spectra, which show a
lowering of only 50-60 cm-' in the C=C stretching fre-
quency (24), and by proton resonance spectra, which do
not differ greatly from those of the free olefins (25).
From the distribution studies (e.g. 17b, 26), evidence
has been adduced for the existence of disilver complexes
of olefins, i.e. (Ag, olefins),+, at relatively high silver ion
concentrations. However, in the chromatographic sys-
tems in which we are interested most of the observed
differences resulting in separation may be satisfactorily
understood on the assumption of 1 :1 complexing.
DEVELOPMENT OF ARGENTATI ON
CHROMATOGRAPHY
Since the first study of silver-olefin complexes by Win-
stein and Lucas (14), such complexing has been used
many times for isolation or purification of olefins by iso-
lation of crystalline complexes or by simple extraction
or liquid-liquid distribution procedures (e.g. 17 e,
and references cited therein). However, the first real
appreciation of the possibilities of argentation as the
basis of sophisticated separatory techniques for lipo-
philic compounds was by Nichols in 1952 (27). I n the
first study of silver complexes of lipid components, he
measured argentation constants of methyl oleate and
methyl elaidate by distribution between isooctane and
aqueous methanol mixtures. He suggested that counter-
current distribution or paper chromatography could be
adapted to separate these two esters and also to separate
saturated and unsaturated esters or olefinic compounds
in general.
The first Chromatographic application of silver-ole-
fin complexing, however, did not follow Nichols' pre-
718 JOURNAL OF LIPID RESEARCH VOLUME 7, 1966

b
y

g
u
e
s
t
,

o
n

S
e
p
t
e
m
b
e
r

1
6
,

2
0
1
4
w
w
w
.
j
l
r
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

diction-this had to wait 9 years for fulfillment-but was
in the realm of GLC. Bradford, Harvey, and Chalkley
(28) found that a saturated solution of silver nitrate
in ethylene glycol, as a stationary phase in GLC, gave
excellent separations of traces of ethane in ethylene.
Tenny (29) clearly illustrated the selectivity of a silver
nitrate-triethylene glycol stationary phase for a series
of normal olefins and the applications of argentation
GLC to separations of volatile olefins were rapidly
extended by other workers (e.g. 30, 31). More recently,
of course, argentation GLC has been used for determina-
tion of argentation constants of large numbers of ethyl-
enic and acetylenic compounds (15, 16) as a simpler
alternative to the classical methods of distribution or
solubilization.
Although very useful for olefins and acetylenes of low
molecular weight, argentation GLC has little relevance
in lipid separations because of the high temperatures
required for GLC of lipid derivatives. Keulemans (30)
stated that the temperature for argentation GLC should
be kept below 40C, because above this temperature
the adducts do not form and the stationary phase is
not stable. This statement was modified by Bednas and
Russell (31), who found that AgNOa-ethylene glycol
and AgNOs-glycerol stationary phases were stable and
could be used satisfactorily at 65 but underwent slow
decomposition above that temperature and could be
used only for short periods at 85C. Even this, however,
is of little value for lipid work.
Nichols prediction was first verified by Dutton, Schol-
field, and J ones (32), who showed that countercurrent
distribution between 0.2 M silver nitrate in 90% meth-
anol and hexane would give excellent separations of
oleate and elaidate and of saturated, mono-, and di-
unsaturated esters. Scholfield and his coworkers have
since extended this method and used it in a variety of
studies (33-35) and de Vries (36) has reported an al-
ternative countercurrent distribution system for the
separation of oleate and elaidate. Countercurrent dis-
tribution with argentation, however, has not been at all
widely used for lipid separations, presumably because
of the tedium of the procedure and because few lab-
oratories have the necessary equipment. It was only the
advent of chromatographic methods based on the silver
ion-double bond interaction which broadcast the po-
tentialities of argentation in lipid separations.
The first reference to argentation chromatography
which this author has been able to trace was the separa-
tion of cis- and trans-5-cyclodecenols by Goering, Clos-
son, and Olson (37). They used a column of silica gel
impregnated with a stationary phase of aqueous silver
nitrate and eluted with benzenelight petroleum and
benzene-ether mixtures. This was, of course, a partition
chromatographic system but similar in effect to the more
recent adsorption systems employing silver nitrate. At
about the same time, reversed-phase partition chro-
matographic methods based on silver complexing and
developed by Wickberg (38) were being used fur sep-
arations of various natural terpenoid materials (39-
42). These methods employed not silver nitrate but silver
fluoroborate in aqueous methanol as mobile phase in
con-iunction either with glass papers impregnated with
hexadecane as stationary phase, or with columns of
polyvinyl chloride powder. Besides the obvious disad-
vantage of using the expensive and dangerous silver
fluoroborate, these procedures resulted in elution of
silver salt with the separated components, which thus
required further purification. However, as separation
methods they were effective and resulted in the detection
and isolation of a number of novel terpenoid com-
pounds (e.g. 38).
True argentation adsorption chromatography, par-
ticularly as applied to lipids, was first described simul-
taneously by Morris (43), who used TLC, and by de
Vries (36), who used columns. Both authors demon-
strated the clear separation of oleate and elaidate, and
of these from saturated and polyunsaturated esters.
Morris, in addition, showed separations of saturated,
ethylenic, and acetylenic hydroxy and epoxy esters and,
by double impregnation with silver nitrate and boric
acid, achieved simultaneous separation of threo and
erythro, saturated and unsaturated dihydroxy esters
(43). De Vries (36), as well as separating methyl esters,
achieved virtually quantitative separations of triglyc-
erides according to total number of double bonds, even
of elaidodipalmitin from oleodipalmitin. He also claimed
quantitative separation of cholesterol from cholestanol.
At the same time Barrett, Dallas, and Padley (44)
described separations of glyceride mixtures on silver
nitrate-impregnated thin layers. These first three short
notes (36, 43, 41) between them indicated the potential-
ities of argentation chromatography in virtually all
important classes of lipid compounds, other than phos-
pholipids and glycolipids, which are only recently being
fractionated by this method.
Only one other significantly different procedure need
be mentioned in this short history of the development of
argentation methods, namely the fractionation of large
amounts of triglycerides by low temperature crystalliza-
tion from acetone-methanol solutions of silver nitrate
described by Gunstone and cowoIkers (45, 46).
SOME PRACTICAL ASPECTS
The practical techniques of argentation chromatography
are, in general, very little different from more conven-
tional chromatographic procedures. Relatively large
samples may be fractionated on columns but analytical
MORRIS Argentation Chromatography 719

b
y

g
u
e
s
t
,

o
n

S
e
p
t
e
m
b
e
r

1
6
,

2
0
1
4
w
w
w
.
j
l
r
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

separations are best performed by TLC and, in many
cases, preparative argentation TLC is the best and most
convenient means of isolating material sufficient for
further analysis and (or) structure determination.
Argentation column chromatography is less satisfactory
than argentation TLC for separations of highly un-
saturated components, such as triglycerides with more
than four double bonds (47), and of compounds with
fairly similar migration behaviour, such as monoacetyl-
enes and cis-monoenes (48).
The adsorbent most commonly impregnated with
silver nitrate has been silicic acid. The disadvantage of
low flow rates of many silicic acid columns may be ob-
viated by using acid-washed Florisil impregnated with
silver nitrate (49, 50). Columns of ion-exchange resins
containing silver ion have also been used (51, 52);
they offer the advantage that polar solvents may be
used without leaching any of the silver ions from the
column. Column chromatographic procedures, of course,
tend to be tedious, hence the much wider use of TLC and
preparative TLC methods, but autoniatic detectors such
as the one described by J ames, Ravenhill, and Scott
(53) in conjunction with argentation column chro-
matography relieve the tedium to some extent.
I n argentation TLC the most common adsorbent by
far is silicic acid. However, alumina impregnated with
silver nitrate has been used a number of times, e.g.
for separations of dinitrophenylhydrazones of various
classes of aliphatic aldehydes and ketones (54, 55) and
for resin acid methyl esters (56), some of which were
unstable on silicic acid. Badings and Wassink showed
that kieselguhr-silver nitrate was also effective for sep-
aration of carbonyl dinitrophenylhydrazone deriva-
tives (57) and, in this excellent paper, they concluded
that besides complexing with C=C bonds the silver ions
also complexed with the C=N bonds of the dinitro-
phenylhydrazones.
Impregnation of thin layers with silver nitrate may be
effected in a number of ways, most commonly by using
an aqueous silver nitrate solution of appropriate con-
centration instead of water for preparing the adsorbent
slurry with which to spread the layers (e.g. 44, 58).
One serious difficulty in preparing plates in this way is
caused by the interaction of the silver nitrate solution
with the metal of conventional plated spreaders. This
interaction has two effects; metallic silver is precipitated
onto the layers being prepared and the spreader itself
becomes rapidly pitted and corroded until, eventually,
it is useless. At least one manufacturer (Desaga, Heidel-
berg, Germany) supplies silver-plated spreaders to
counteract this problem or, alternatively, a spreader
made of anodized aluminum, such as the excellent
spreader produced by Quickfit and Quartz Ltd., Stone,
Staffs., England that we use, is impervious to silver
nitrate. A third possibility, which we use for small
plates ( 3l / 4 inches square), is to construct a spreader
from some suitable plastic sheet. As an alternative to this
direct method of making impregnated layers, thin-layer
plates already prepared may be wholly or partly impreg-
nated by spraying with an aqueous or methanolic solution
of silver nitrate (43) or by development with an aqueous
solution of silver nitrate (59). These last two procedures
permit comparisons of migration behavior on normal and
impregnated adsorbents on a single plate.
The level of silver nitrate impregnation recommended
has varied from 30-40y0 (60) down to 3y0 (61). Using
sample mixtures of saturated, monoenoic, and dienoic
fatty acid methyl esters, wax esters, and cholesterol
esters on a whole series of plates of 0.5-30y0 silver nitrate
content (i.e. silver nitrate: silicic acid ranged from 0.5:
99.5 through 30:70, w/w), we found that there was no
improvement in separations above 2y0 silver nitrate
(58, 1). A similar conclusion was reached by Klein,
Knight, and Szczepanik (62) for steroids and by Stahl
and Vollman (61) for terpenoid alcohols. The last
authors arrived at their conclusion very elegantly by
means of a layer having a linear gradient of impregna-
tion with silver nitrate, prepared with Stahls GM-
spreader (63), as illustrated in Fig. 1. I n this illustration
a level of about 1.5% silver nitrate is seen to be necessary
for separation of nerolidol from geraniol, but at this level
of impregnation guaiol and borneol are inseparable.
At about 2% impregnation all components are separated
and there is no further improvement of separation with
increasing silver nitrate concentration. Note the cross-
over of guaiol and borneol, the former being less mobile
than the latter at low levels of impregnation and more
mobile at higher levels. Stahl considered that 3% im-
pregnation was the optimum and most economical level
of impregnation, whereas we normally use 5%. The
only case where we have found higher levels of impregna-
tion (loyo or more) to be advantageous is for separations
of positional isomers of monoenoic fatty acid esters
(64). We do not know why this should be, but it may
be that disilver-olefin complexes have some role in these
separations. One disadvantage of higher levels of im-
pregnation for TLC, apart from the expense, is that de-
tection by corrosive reagents becomes progressively
more difficult.
Impregnated plates are activated in the usual way;
we heat at llOC for 30 min for thin layers or for 60
min for thicker layers (1 mm). Plates should then be
stored in sealed glass containers because laboratory
fumes frequently cause rapid deterioration in the selec-
tivity of impregnated layers. We store the plates over
saturated calcium chloride solution (relative humidity
ca. 300/0), thereby ensuring reproducible results even
after several weeks.
720 J OIJ RNAL OF LIPID RESEARCH VOLUME 7, 1966

b
y

g
u
e
s
t
,

o
n

S
e
p
t
e
m
b
e
r

1
6
,

2
0
1
4
w
w
w
.
j
l
r
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

Fro. 1. Gradient layer thin-layer chromatogram of iiiono- and scsquitcrpcnic alcohols demonstrating the effect
of increasing levels of impregnation with silver nitrate. Ihc adsorbent was Silica Gel 1-1 (Mcrck) with a linear
gradient of silver nitrate froiii 0 to 2.5;; and the solvent was incthvlenc chloride-chloroform-ethyl acctatr-n-
propanol 50: 50: 5: 5. Samplrs: I , nerolidol; 2, geraniol; 3, nerol; 4, guaiol; 5, borneol; 6, cedrol.
Reproduced from Stahl and Vollinann (61 )with the perinision of the authors and Pergainon Pres Ltd.
The choice of visualization reagents is somewhat re-
stricted by the presence of silver nitrate; for example io-
dine vapors are no longer suitable. However, corrosive
agents such as aqueous sulfuric acid (43), phosphomolyb-
dic acid in ethanol (65), chlorosulfonic acid-acetic acid
(66), or phosphoric acid (67), all followed by heating at
a suitable temperature, are satisfactory as general de-
tection reagents. Visualization under ultraviolet light
after spraying with dichlorofluorescein (43, 58) or
dibromofluorescein (44, 67), or in daylight after spray-
ing with water (65) is also suitable for general detection
and, of course, for preparative work. Quantitative anal-
ysis of separated components may be effected by any
of the procedures normally used in conjunction with TLC.
Several other analytical methods for lipids deserve
mention in this section. Wood and Snyder (68) very
recently described a modified argentation TL C system
wherein the plates were prepared with a solution of
silver nitrate in ammonia instead of i n water, and the
active species in coinplexing was considered to be the
diamminesilver ion Ag(NHa)*+. These plates were
reported to give better separations of fatty acid methyl
esters, to retain their resolving power much longer, and
to involve less corrosion of conventional spreaders, claims
with which we concur.
A number of reversed-phase partition methods in-
volving silver ions have also been described. Apart from
the methods developed by IVickberg (38) employing
silver fluoroborate i n the mobile phase, which have been
iiientioned already, Vereshchagin (69) and Paulose (70)
have described separations of lipids by paper chroinatog-
raphy and TL C respectively with a nonpolar stationary
phase and silver nitrate i n the inobile phase. These
procedures separate according to chain Ienqth and degree
of iinsaturation siinultaneously and avoid the critical
groups which occur i n conventional reversed-phase
chromatography. Paulose (70) impregnated his layers
with silicone oil and then again with silver nitrate by
spraying, but this latter step was almost certainly un-
necessary because it was clearly the silver ions in the
mobile phase which effected the separations.
One thing that has surprised this author is that no one,
apparently, has utilized the treniendous solubility of
silver nitrate in acetonitrile as the basis of a partition
chromatography or countercurrent distribution pro-
ced tire.
SEPARATI OSS BY
ARGESTATI ON METHODS
I n the following sections, classified according to cotn-
pound types, the various possibilities of lipid separations
MORRIS Argentation Chromaloqraphy 721

b
y

g
u
e
s
t
,

o
n

S
e
p
t
e
m
b
e
r

1
6
,

2
0
1
4
w
w
w
.
j
l
r
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

by argentation chromatography are briefly reviewed.
There is some discussion of suitable combinations of
atyentation methods with other separatory and analyt-
ical methods and of some of the ways in which these
methods have already been employed in research on
lipids.
Fait?/ Aci ds
The most obvious, and probably most common applica-
tion of argentation chromatoqraphy is in separation of a
mixture into fractions of diffcrinq numbers of double
bonds (usually ci s), and many such fractionations of fatty
acid mixtures have been reported. separations of fatty
acid methyl ester mixtures i n this way have been car-
ried out, on a relatively larqc scale, by coluinn chro-
matography on silicic acid (e.q. 36, 48, 60, 71), Florisil
(49, 50), or ion-cxchanqe resins (51, 52) iinpregnated
with silver nitrate. On an analytical or sniallcr prepara-
tive scale, TLC 011 silicic acid iinpregnated with silver
nitrate (c.g. 43, 72, 73) or with ainmoniacal silver ni-
trate (68) is most convenient and the deqree of sep-
aration obtained is illustrated i n Fiq. 2. These separations
have been carried out qencrally with methyl ester mix-
tiires, but free acids may be similarly separated if a
little formic or acetic acid is added to the developing
solvent to suppress dissociation. Alcohols and other sim-
ple aliphatic coinpounds are also easily separated. Some
excellent separations of aldehydes and ketones as their
dinitrophenylhydrazone derivatives have also been
carried out by TLC on silver nitrate-inipregnated alu-
mina (54, 55) and kieselqthr (57).
As with all forms of chroniatoqraphy, arqentation
chroniatoqraphy is most effectively employed in con-
junction with other types of chroniatoqraphy. The fact
that it separates basically accordinq to deqrec and type of
unsaturation, with little i f any separation of differing
chain lenqths, makes it particiilarly suitable foi com-
bination with liquid-liquid and gas-liquid methods.
Rerqelson, Dyatlovitskaya, and Voronkova (74) de-
scribed two-diinensional TLC of esters, reversed-phase
TL C in the first and arqentation TL C i n the second
dimensions, for coriiplctc separation according to both
chain lenqth and deqrcc of unsatiiration. The reverse
sequence was used by Radinqs and Massink (57) for
carbonyl dinitrophcnylhydrazonc derivatives. Vere-
shchaqin (69) and Paulose (70) obtained the sainc type of
complete separation in one dimcnsion, on paper and
by TL C respectively, by rcvciscd-phase chroinatoqraphy
employinq a mobile phasc containinq silver nitrate. This
same type of separation, by chain lenqth and iinsatwa-
tion, is of coiirse the result of countercrirrent distribution
with silver ions in the polar phase (32,34)
A rather exotic variant of coinbined chromatographic
techniqrics was recently described (75), wherein the
. * -
A B C D E F C
FIG. 2. Thin-laycr chroiiiatoqram of fatty acid incthyl cstrrs and
siinplc lipids on Silica Grl G iniprcgnatcd with silver nitrate
(5%, w/w). Dcvclopinq solvcnt was dirthyl cthrr-liqht pctrolcurn
5 : 3 5 and spots wcrr located by sprayinq with 50; IIzSOC and
charring. S,unplcs: A, inethyl stearate; B, inrthvl claidatc; C,
incthyl oleatc; D, mixture of stcaratr, olcatc, and linolratc; E,
iiirthyl cstcrs froin fccalith lipids; F, inisturc of cholcsteryl
straratc, cholcstcryl olcatc, and cholrstcryl linolratc; C, spcrin
oil, i.c., a wax rstcr inixturc.
Kcproduced froin Morris (2) with thc permission of John
\Vilry k Sons Ltd.
first dimension of chroinatography was GLC, the
components emerging from the column being eluted on
to the edge of a logarithmically travelling thin-layer
platr impregnated with silver nitrate, which was de-
veloped to provide the second diinension. \2hile ad-
niirinq the elegance of this method, this author has some
doubts on the returns in practical utility which may be
expected for the investment in the rather elaborate
equipnicnt and procedures necessary. The much simpler
system of preliminary fractionation by preparative
argrntation TLC followed by GLC of the individual
unsaturation classes is perfectly adequate for identifica-
tion of most compounds and is niorc suitablc for detec-
tion and identification of minor components and for
quantitative analysis of complex mixtures. This latter
approach has been used by many workers (e.g. 1, 48,
76, 78) and is excniplified i n Fig. 3.
After argentation chroniatography, each iinsaturation
class may be fractionated according to chain length by

b
y

g
u
e
s
t
,

o
n

S
e
p
t
e
m
b
e
r

1
6
,

2
0
1
4
w
w
w
.
j
l
r
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

L
20:o
3
18: 3 2 0 3
v \
A 6 16:4
FIG. 3. Reproduction of the GLC curves (obtained on an ethylene
glycol adipate polyester stationary phase) of sardine oil total
mixed esters (Total), and the saturated (0), monoene ( 7) , diene
( Z) , triene (3), tetraene (4), pentaene (5), and hexaene ( 6) ester
fractions. These fractions were isolated by preparative TLC on
Silica Gel G impregnated with silver nitrate (5%, w/w) developed
with diethyl ether-light petroleum 5 : 95 for the 0, 1, and 2 frac-
tions, and methanol-diethyl ether 1 : 9 for the 3, 4, 5, and 6 frac-
tions. Note the separations of positional isomers of dienoic and
trienoic esters.
Reproduced from Morris (1 ).
preparative GLC and the structure of each compound
or mixture so obtained determined by permanganate-
periodate oxidation (e.g. 78) or by reductive ozonolysis
(e.g. 73).
These and other procedures have assisted in the de-
tection and characterization of a whole range of novel
cis-ethylenic and acetylenic acids from natural sources
(a number of trans-ethylenic acids have also been dis-
covered but will be described later). Thus 4,7,10,13-
eicosatetraenoic acid was detected in rat liver phos-
pholipids (73) and several odd-number and isomeric
acids were shown to be minor components of a number of
seed oils (76, 77). Alpha- and y-linolenic acids and 6,9,-
12,15-octadecatetraenoic acid were shown to occur in
several Boraginaceae seed oils (79) and all-cis 5,9,12-
octadecatrienoic acid was demonstrated in Xeranthemum
annumseed oil (80). Stearolic acid was shown t o be pres-
ent in la number of Santalaceae seed oils (81, 82) and
more complex acetylenic acid mixtures from the seeds
of Onguekoagore (83, 84) and Acanthosyris spinescens (85)
were separated into fractions with and without terminal
ethylenic bonds and the individual components of these
fractions then identified.
So far in this section, we have only discussed separa-
tions on the basis of degree of unsaturation. One of the
most useful attributes of argentation methods is that they
can separate compounds with nonconjugated unsatura-
tion according to the geometry of the double bonds.
I n this they have an advantage over methods involving
mercuric salt adduction, which do not separate ci s-
and trans-ethylenic isomers. Separations of cis- and trans-
monoenoic esters are readily achieved by argentation
adsorption chromatography in columns (e.g. 36, 60)
or on thin layers (e.g. 43, 64), by partition methods
on thin layers (70, 74), and by countercurrent distribu-
tion (32, 34, 36). The separation of elaidate from oleate
and stearate by TLC is shown in Figs. 2 and 4 and an
example of one application of this type of separation, the
separation of the cis and trans monoenes from fecalith
mixed esters, which were then analyzed separately by
GLC and shown to be very different in chain-length dis-
tribution (86), is also illustrated in Fig. 2 (sample E).
Geometrical isomers of unsaturated aldehyde dinitro-
phenylhydrazones are similarly easily resolved by TLC
(57, 87). Separations of the various geometric isomers of
polyenoic esters are also readily achieved by counter-
current distribution (e.g. 34) or TLC (e.g. 34, 88, 89).
The ability to fractionate ester mixtures according to
the geometry as well as the degree of their unsaturation
has been particularly useful in studies of the mechanism
and intermediate products of hydrogenation. Both hetero-
geneous (e.g. 33, 35, 90) and homogeneous (e.g. 91, 92)
catalytic hydrogenation of unsaturated esters have been
studied, and J ardine and McQuillin (93) have recently
compared rates of catalytic hydrogenation of various
ethylenes and acetylenes (not fatty acids, however) with
their retention on silicic acid-silver nitrate thin layers
and have found some parallelism. Partial reduction of
polyunsaturated esters with hydrazine, which causes no
positional or geometrical isomerization of unreduced
double bonds, has also been studied with the help of
argentation TLC (e.g. 94). Indeed, partial reduction
with hydrazine, isolation of the cis or trans monoenoic
products by argentation TLC, and determination of
their structures by GLC after oxidation or ozonolysis
comprise the most powerful combination of techniques
yet devised for unequivocal elucidation of the structures
of polyunsaturated fatty acids. This approach has already
Moms Argentation Chromatography 723

b
y

g
u
e
s
t
,

o
n

S
e
p
t
e
m
b
e
r

1
6
,

2
0
1
4
w
w
w
.
j
l
r
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

been iised a number of times on novel acids from various
seed oils; for example all-cis 5,9,12-octadecatrienoic
acid (go), cis-9, tram-11, trans-13, cis-1 5-octadecatetra-
enoic acid (95), cis-9, trans-12-octadecadienoic acid
(96), and ~rons-3, cis-9, ci s-1 2-octadecatrienoic acid
(97) ; and has very recently been written up as a defined
and general procedure by Privett and Nickell (98).
I n some cases, for example ci s, frons-linoleate (96) and a
series of trans-Aa-acids (97), the novel acids were de-
tected in the first place and were isolated by argentation
TLC prior to final elucidation of their stnictiires by the
a bove procedure.
Resides effecting separations according to degree of
unsatiiration and to geoiiietry of double bonds, argenta-
tion chroinatoqraphy has a third ma-ior attribute, namely
the ability to separate suitable positional isomers of
unsaturated fatty acids. Scholfield, J ones, Rutterfield,
and I htton by countercurrent distribution (34) and de
Vries and J urriens by TLC (88) showed that with dienoic
esters the effect of silver complexing increased with in-
creasinq separation of the two double bonds. Thus, 9,11-,
9,12-, and 9,15-octadecadienoates could be readily
separated and, on this basis, a number of positional
isomers of linoleic acid were detected in butter fat by ar-
gentation TLC (99). De Vries and J urriens also demon-
strated (88) the separation of the 6-, 9-, and 1Zocta-
decenoate isomers by argentation TLC using a benzene-
light petroleum mixture as solvent. Similar separations
of the 7-, 9-, and 11-octadecenoates and of the 9- and
11-isoiners were reported by Bergelson et al. (74),
and by Lees and Korn (89), respectively, using mixtures
of ethers and liqht petroleum fractions, and Vereshchagin
(69) separated the 6- and 9-octadecenoates by his
reversed-phase paper system. We have been unable to
obtain adequate, reproducible separations of isomeric
nionoenoic esters by argentation TLC with the solvents
and conditions described by these workers. We have de-
vised a procedure (64) involvinq double or triple de-
velopinent with toluene at -15OC or -25"C, which
consistently gives quantitative separation of suitable mix-
tures of isomeric monoenoic esters, as illustrated in Fig.
4. This system, incidentally, gives much better separations
of mixtures containing acetylenic acids than are obtained
with the more conventional ether-light petroleum sys-
tems (81,100).
Although all of the actadecenoate isomers have not
been examined, weconsider (64) that both the cis and the
trans series will conform to a sinusoidal type of curve
siinilar to that produced by normal TLC of positionally
isomeric substituted esters (101, 102) ; i.e. decreasing
mobilities from A2 t o about A , then progressively in-
creasing mobilities to about AI?, and decreasing mobilities
aqain to A". Positionally isomeric unsaturated aldehyde
FIG. 4. Thin-layer chroinatoqrani of soinc fatty acid inethyl
rstcrs on Silica Gel G iiiiprcgnated with silver nitrate (lo';,
w/w). The plate was developed three tiincswith tolueneat -15OC
and spots were located by spraying with chlorosulfonic acid-acetic
acid 1 :2 and charring. Saiiiplcs: A, methyl claitlatc (upper) and
incthyl oleate (l owrr); R, inethyl cis-vaccenatc (All, upper),
nirthyl oleate (Ag, middle) and methyl pctrosrlinate (A#, Iowa);
C, rat liver mixed esters (small proportion of Ai 1- 18: l not visible
on reproduction); I ) , parsley seed oil mixed inethyl esters (separa-
tion of A9- and A6-18:l isomers); E, methyl stcarolate; F, Exo-
carpus cu/nessi/ormis seed oil mixed inethyl esters. In this last sample
( F) the coinponents above oleate, the iiiajor constituent, arc
stcarolate (a small amount), santalbatc (octadcc-9-yn-1 l-cnoate,
thc other major constituent), a novel furanoid acid ester (see trxt),
and saturated esters (sinal1 amount).
derivatives have also been separated by argentation
TLC (e.g.57,87).
ArKentation chroinatography is also very suitable for
separation of most substituted fatty acids according to
degree and type of unsaturation. This is particularly
valuable because the conditions required for mercuric
adduct formation degrade some acid-labile compounds,
such as epoxy esters, and GLC is also unsuitable for a
number of hydroxy esters (e.g. 103,104).
A considerable range of epoxy, halohydroxy, hydroxy,
and dihydroxy esters have been separated by argentation
TLC according to degree, type, and position of unsatura-
tion by Morris (43) and Morris and Wharry (101).
Dihydroxy esters were separated not only on the basis
of unsaturation but also according to the threo or q t h o
configuration of the glycol group by double impregna-
tion with silver nitrate and boric acid (43). In com-
bination with normal adsorption TLC and GLC, which
721. JOURNAL OF h P I D RESEARCH VOLUME 7. 1966

b
y

g
u
e
s
t
,

o
n

S
e
p
t
e
m
b
e
r

1
6
,

2
0
1
4
w
w
w
.
j
l
r
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

effect remarkable separations of positionally isomeric
substituted esters (101,102,104), argentation chromatog-
raphy is a great aid in elucidation of structures of novel
oxygenated acids. By these means the structures of the
acetylenic hydroxy acids of isano oil (83, 84), an epoxy
acid in Aster aZpinus seed oil (97), and a whole series of
previously uncharacterized a-hydroxy acids from brain
cerebrosides (48, 105) have been determined. The study
of that currently very exciting class of oxygenated fatty
acid derivatives, the prostaglandins, has also been greatly
facilitated by a combination of TLC and argentation
TLC (e.g. 65, 106-109). The separation of all of the
known prostglandins described by Grten and Samuels-
son (65) provides an excellent example of the scope and
selectivity of this combination of methods. We have very
recently detected and isolated, by argentation TLC,
a highly unusual oxygenated fatty acid ester from the
mixed esters derived from Exocarpus cupressiformis seed
oil (Fig. 4, sample F) and proved it to be methyl 8-(5-
hexylfuryl-2)-octanoate (100).
Nonoxygenated substituted acids are also separable
by these methods. A whole range of branched-chain
acids from skin lipids were characterized by GLC after
preliminary fractionation of the mixed esters according to
unsaturation on a silicic acid-silver nitrate column (1 10).
Cyclopropanoid esters may be separated from unsatu-
rated esters (e.g. l l l ), but an early claim that cyclo-
propenoid esters could be separated from normal sat-
urated and unsaturated esters by argentation TLC (112)
seems to be generally refitted by others (e.g. 5). The break-
down of cyclopropenoid esters is presumably due to the
ready reaction of such compounds with silver nitrate
(cf. 113). By formation of methyl mercaptan derivatives
of cyclopropenoid esters, however, these compounds
are readily separated from unsubstituted and cyclo-
propanoid esters by argentation TL C (114).
Neutral Lipidp
Argentation chromatography of intact lipid classes,
particularly of glycerides, has been applied much more
widely and usefully in the short time since its inception
than has reversed-phase partition chromatography.
This is because it is more convenient, more definitive,
and more suitable for isolation of sufficient quantities
of individual fractions for further study, e.g. by lipolysis.
Cholesterol esters from blood serum have been quan-
titatively separated by argentation TLC into seven frac-
tions having zero to six double bonds (58) and the less
complex sterol esters from skin have been fractionated by
column and thin-layer argentation chromatography
(115). Separation of a simple cholesterol ester mixture
is illustrated in Fig. 2 and the fractionation of a wax
ester mixture, sperm oil, is also shown in the same figure.
The obviously composite spots of the monoenoic and
dienoic components of sperm oil was shown by GLC
to be due, largely, to chain-length variations (1, 2).
The wax esters from skin have also been resolved in this
way (115) and the isolation of several homologous series
of hydrocarbons from various plant waxes by argenta-
tion chromatography (116) should also be mentioned
here.
I n one of the first publications on argentation TLC
(44) separations of diglycerides and monoglycerides
according to unsaturation were described and this pro-
cedure has very recently been extended to the equally
easy fractionation of glycerol ethers (1 17). However,
the vast majority of neutral lipid separations by silver
ion complexing have been of triglycerides.
Of the first three papers on argentation chrornatog-
raphy of lipids, two described separations of triglycerides,
on columns (36) and on thin layers (44). Columns of
silicic acid impregnated with silver nitrate have been
used on many occasions (e.g. 45-47, 118-120) for pre-
parative fractionation of natural triglyceride mixtures
but are not very satisfactory with glycerides containing
more than four cis double bonds (47). Gunstone and co-
workers (45, 46), however, partially circumvented this
difficulty by preliminary fractionation of relatively large
amounts of triglyceride mixtures by low temperature
crystallization in the presence of silver ions. The simpler
mixtures thus obtained were then separable by column
chromatography. Argentation TLC, however, has proved
more popular even for preparative work because of its
greater convenience and the superior separations ob-
tained, and it has been used by many authors in studies
of natural triglyceride compositions (e.g. 67, 121-1 32).
Quantification has been achieved in a variety of ways:
by densitometry after charring (67), by glycerol deter-
mination after hydrolysis (122, 123), by colorimetry after
reaction with hydroxamic acid and ferric ions (124)
or with chromotropic acid (129), or by GLC of the com-
ponent fatty acids after addition of a suitable internal
standard (125, 131). Lipase hydrolysis of the separated
triglyceride fractions was used in most cases to obtain
still more information on glyceride compositions.
An example (not particularly good) of argentation
TLC of some natural triglyceride mixtures is illustrated
in Fig. 5. I t is clear from this that some triglycerides
having the same total number of double bonds are clearly
separated from each other. Gunstone and Padley (125)
determined the order of complexing powers of triglyc-
erides, with from nine to zero double bonds and without
considering positional isomers, to be as follows (3 =
linolenic, 2 =linoleic, 1 =oleic, and 0 =palmitic and/
or stearic): 333, 332, 331, 330, 322, 321, 320, 311, 222,
310, 221, 300, 220, 211, 210, 110, 100, 000. They have
found it useful to assign arbitrary values for the com-
plexing power of each acyl chain, viz. saturated =0,
MORRIS Argentation Chromatography 725

b
y

g
u
e
s
t
,

o
n

S
e
p
t
e
m
b
e
r

1
6
,

2
0
1
4
w
w
w
.
j
l
r
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

Front
A B
c o
FIG. 5. Thin-layer chromatopmn of natural triglyceride mix-
turn on Silica Ccl G iinpregn;itcd with silvcr nitrate (S:;, w/w).
The solvent was isopropanol-chloroform 1.5: 98.5 and spots were
locatcd by sprayinq with 50"; Il?SO, and charring. Samples:
A, palm oil; R, olive oil; C, groundnut oil; D, cottonseed oil.
The nuinbcrs rcprcscnt total numbers of double bonds in each
triqlyccridc molccu IC.
Reproduced from Morris (2) with the pcrinission of John \%ley
& Sons I d .
oleic =1, linoleic =(2 + a) , and linolenic =(4 +
4a), where a is some fraction less than unity. Thus the
triglyceride 330, with six double bonds, has a complexing
power of (8 + 8a) which is greater than the value of
(8 + 6.) for the triglyceride 322, which has seven double
bonds. Positional isomerism may result in further separa-
tion of some of the classes listed above (44).
Even greater selectivity of separation is, of course, ob-
tained by cornplenientary use of argentation chromatog-
raphy with some other chromatographic method. Thus,
Cubero and Mangold (59) have combined it with nor-
mal adsorption TLC to separate triglycerides, free acids,
and sterols as classes and then to resolve theni according
to unsaturation in the second dimension. Blank and
Privett (128) also coinbined it with normal adsorption
TLC to determine the cornposition of inilk fat tri-
glycerides. Kaufrnann and Wessels, on the other hand,
coiiibined it with rcvcrsed-phase partition TLC to
separate, i n two dirnensions, virtually all the glyccrides
of sunflower oil (127), while Vercschaqin used the same
coinbination siinriltaneorisly on paper (69). Pcrhaps the
most promising coinbination for triqlyccridc scparations,
howcvcr, is of argentation chroinatoqraphy with GLC
of thc intact glyccridc fractions so obtaincd (129, 130)
or of the azclao-glyccridcs produccd from them by
oxidativc clcavaqc (47).
Thc miich grcater precision of analysis now possible
by thcsc incthods has lcd Privctt and coworkers (131,
132) to call into question one of the basic assuinptions of
Vandcr \Val's 1,3-randoin, 2-random hypothesis of
glyccride distribution, namely that thc fatty acids i n the
2-position arc randomly distributed rclativc to the 1,3-
positions. Morris (133) has devised a procedure, based
on separation by argentation TLC of derivatives of the
diglyccrides produced froin pure triglyccrides by lipol-
ysis, to dcrnonstratc optical asyinmctry i n natural tri-
glycerides for the first tiine. This, of coursc, is counter to
the other assumption of the 1,3-randoin, 2-random theory
that the 1- and 3-positions are eqriivalcnt.
Polrrr Lipid9
Phospholipids and glycolipids arc inore difficult to
scparate according to unsaturation by argentation
chromatography than arc neutral lipids and, so far at
least, relatively little work in this direction has been
done. Kaufinann, ~4"cl s, and Bondopadhyaya (1 34)
claimed to have separatcd soya lecithin and egg lecithin
into nine and seven fractions, respectivcly, by argenta-
tion TLC. Howevcr, using their solvent system (chloro-
form-diethyl ether-acetic acid 97 :2.3:0.5), we have been
unable to reproduce their separations or even to effect
migration of a pure saturated lecithin froin the origin.
More convincing separations of a nuinher of natural
lecithins have recently been reported by Arvidson (1 35)
and are illustrated in Fig. 6. Less complete fractiona-
tions of lecithins from a number of animal tissues (136,
137) and of phosphatidyl glycerols froin photosynthetic
tissues (138, 139) have been described by van Deenen
and his collaborators.
A more complete fractionation and analysis of the
molecular species of various phospholipid classes is
possible if recovery of the intact phospholipid fractions is
not required (140). This is effected by first converting
the relevant polar lipid class to the corresponding di-
glycerides; these, or derivatives of them, being neutral
lipids, are easily separable into individual unsaturation
classes by argentation TLC. This approach was pioneered
726 JOURNAL OF LIPID RESEARCH VOLUME 7, 1966

b
y

g
u
e
s
t
,

o
n

S
e
p
t
e
m
b
e
r

1
6
,

2
0
1
4
w
w
w
.
j
l
r
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

a 6
FIG. 6. Thin-layer chromatograms of lecithins on Silica Gel H impregnated with silver nitrate
(ca. 20%, w/w). The solvent was chloroform-methanol -water 65: 25: 4, plate a having been
activated at 175OC for 5 hr and plate 6 at 180C for 24 hr. Detection was by charring after
spraying with 50% H2S04. Samples: A, hydrogenated egg lecithin; R, egg lecithin; C, rab-
bit liver lecithin; D, rat liver lecithin; I?, pig liver lecithin; F, bovine liver lecithin. The
numbers represent the total numbers of double bonds in each lecithin molecule.
Reproduced from Arvidson (135) with the permission of the author.
by Renkonen (140, 141) and has been applied success-
fully by him and by van Deenen and his school (136-
139) for the analysis of a number of polar lipids from
various sources. Whereas Renkonen initially produced
acetoxy diglycerides from the phospholipids by acetolysis
(141), van Deenen (136-139) utilized enzymatic cleav-
age with phospholipase C. Very recently, Renkonen
(142) has reported that the acetolysis procedure induces
some intramolecular rearrangement of the fatty acids
on the diglycerides and, therefore, he has also recom-
mended the use of phospholipase C for this cleavage.
Sterols and Other Terpenoid Compounds
I n his first paper on argentation column chromatography
(36), de Vries claimed quantitative separation of choles-
terol from cholestanol. Avigan, Goodman, and Steinberg
(143) showed that silver nitrate-impregnated plates
gave excellent separation of 5,7-dienic sterols from
saturated and monounsaturated sterols. Morris (1-3)
also noted the strong complexing of 5,7-dienes on argen-
tation TLC and showed that whereas A5-stenols were
retained strongly enough to be clearly separated from
cholestanol, as illustrated in Fig. 7, A-stenols were not
(cf. 144). Free sterols differing only in side-chain un-
saturation have very similar mobilities on silver nitrate-
impregnated plates but if less polar derivatives, such as
acetates, are made separation on the basis of side-chain
unsaturation is possible (e.g. 4, 143).
Several similar types of separations of small numbers
of sterols by argentation TLC have been reported (e.g.
145-147) but the largest ranges of sterols so far examined,
mostly as their acetates, were described by Copius
Peereboom (148, 149) and by Klein et al. (62). These
three excellent papers show clearly the great influence
of the molecular environment around centers of unsatura-
tion on the degree to which they participate in complex-
ing. They also, incidentally, demonstrate the consider-
able potential of argentation TLC, particularly in con-
junction with reversed-phase partition (148, 149) or
adsorption (62) chromatography, for analytical and
structural studies of sterols. Whereas most of the authors
cited above have assumed that silver ion-double bond
complexing provides the selectivity whereby these sep-
arations of various sterols are attained, Klein et al.
(62) have provided fairly convincing evidence to suggest
that silver-olefin complexing may be less effective in
TLC of sterols than are geometrical changes i n the ad-
sorbent pore diameter induced by the impregnation.
Clearly there are more complex and structurally specific
effects in argentation TLC of sterols and other alicyclic
MORRIS Argentation Chroma1ograph.y 727

b
y

g
u
e
s
t
,

o
n

S
e
p
t
e
m
b
e
r

1
6
,

2
0
1
4
w
w
w
.
j
l
r
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

* . * * I
A B C D E F G H
0 . 0 . . , ! i
E F G H
FIG. 7. Thin-layer chromatograms of sterols on Silica Gel C and on Silica Gel G impregnated with
silver nitrate (5$,, w/w). The solvent wascommercial chloroform and spots were located by spraying
with 50:'; I 12SOr and charrinq. Samples: A, coprostanol; R, cholestanol; C, cholcsterol; I ) , A7-chol-
estenol; E, 7-dehydrocholrsteroI ; I*., ergosterol; G, @-sitosterol; If, stiqinasterol.
Kcproducrtl froin Morris (2) with thr prrmision of.lohn IViIrv R Sons L!d.
coinpounds than there are with fatty acids and thcir
derivatives, and more basic comparative work will have
to be done before the full potentialities of these methods
are realized.
A considerable nuinbcr of papers have described
argentation-chroinatographic separations of a wide
variety of terpenoid compounds other than sterols.
Although these various compounds are not lipids, at
least not under this author's classification of lipids, they
are certainly lipophilic and deserve a brief summary
here. The first argentation separations of terpenoid
materials employed the reversed-phase systems with sil-
ver fluoroborate in the mobile phase devised by Wick-
berg (38). These procedures were used for separations
of various sesquiterpenic hydrocarbons and resin acid
methyl esters (38-42). Sesquiterpcnic hydrocarbons were
the first of this class to be studied by adsorption TL C
on silver nitrate-impregnated silicic acid by Gupta and
Dev (66), more recent work having been reported by
Czech workers using coluinns (150) and TLC (151, 152).
The range of terpenoid compounds studied was extended
to diterpenic hydrocarbons, esters, aldehydes, ketones,
and alcohols by Xorin and \Vestfelt (153), who combined
argentation TLC with GLC of these compounds. Ar-
gentation TL C is particularly suitable for separation
and identification of resin acid methyl esters (153, 56),
some of which are subject to alteration on GLC. Mono-,
sesqui-, and diterpenic alcohols were studied by Stahl
and Vollman (61), who used argentation TLC in con-
iunction with adsorption TLC and reversed-phase
partition TLC, and Sano and Martelli (154) described
separations of some pairs of allylic-propcnylic isomers.
More complex structures have also been separated by
argcntation TLC, such as tetracyclic triterpenes (155),
steroidal alkaloids and sapogenins (1 56), and ethynyl
steroids from other stcroids (157). I n this last casc, how-
ever, it is not silver ion coniplexing which produces the
separations but silver-acetylene salt formation.
CONCLUSI OSS ASD PROSPECTS
That argentation chroinatqraphy is an extremely
valuable addition to the tcchnical arsenal of the lipid
chemist or biochemist should now be abundantly clear.
Argentation TL C has the greatest versatility and, there-
fore, is of most value; but paper and column chromatog-
raphy and counterciirrcnt distribution employing silver
ions all have their uses, and the last two procedures are
valuable when relatively large amounts must be frac-
tionated. For the most part, argentation chromatography
has siiperscded the older methods based on mercuric
acetate adducts, for example, but there are still occasions
when these older methods may be desirable or necessary.
Argcntation methods, of course, are generally most ef-
fective when used in conjunction with other separatory
and analytical procedures and this aspect has been em-
phasized throughout this review.
The applications of argentation Chromatography in
lipid chemistry and analysis have been summarized
fairly comprehensively during the course of this review.
Its value in analysis of complex ester mixtures, in de-
128 JOURNAL OF LIPID RESEARCH VOLUME 7, 1966

b
y

g
u
e
s
t
,

o
n

S
e
p
t
e
m
b
e
r

1
6
,

2
0
1
4
w
w
w
.
j
l
r
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

tection and structure elucidation of unusual fatty acids
and in determination of the fatty acid distribution in
triglycerides and polar lipids, for example, should by
now be self-evident. However, relatively little has been
said of applications of argentation chromatography in
biochemical investigations.
Until now, the applications of argentation chromatog-
raphy in biochemical studies have mostly been very
marginal; i.e., simply for isolation or analysis of pre-
cursors or products. Thus, it has been valuable in isolating
and purifying synthetic labeled precursors such as elaidic
acid (158) or the isomeric cis- and trans-2- and -3-hexa-
decenoates (159) and -dodecenoates (160) and in iso-
lating and identifying the products of biological reac-
tions (e.g. 159-163). We use argentation TLC routinely
in our laboratory, in conjunction with preparative GLC,
for isolation of individual acids produced biologically
so as to determine the position of the 14C-label by hydro-
genation, chemical a-oxidation, and radio GLC (e.g.
162, 163). I n this context, it may be pertinent to point
out that unsaturated fatty acid esters tritiated at the
position of unsaturation are retained slightly less strongly
on silver-impregnated adsorbents than the corresponding
unlabeled or 14C-labeled esters (164). A similar isotope
effect of deuterated olefins had previously been demon-
strated by argentation GLC (165) and these findings
indicate that some care is required in chromatography
of compounds labeled with deuterium or tritium. Other
biochemical investigations which have been greatly
facilitated by argentation TLC are studies of the bio-
hydrogenation of unsaturated fatty acids in the ovine
digestive tract (166) and by rumen microorganisms (167)
and the excellent work on biosynthesis of prostaglandins
(e.g. 106, 107, 109).
However, in all of these studies summarized above,
argentation chromatography has been of only incidental
importance as a convenient means of isolating the fatty
acid products of biological reactions. I n the future it
will undoubtedly assume far greater importance in studies
of the metabolism and physiological significance of intact
lipids by becoming a basic feature of experimental de-
sign. I t will be particularly useful for studies of absorp-
tion, transport, and interconversions of lipids that use
double labeling techniques. As an example, the long-
standing question of whether or not intact triglycerides
are absorbed in the gastrointestinal tract (cf. 168)
could be readily resolved by the complementary use of
argentation TLC and a doubly labeled substrate. Thus,
a mixture of, for example, (aa-14C2, /3-3H)-triolein with
a large excess of unlabeled tripalmitin could be used as
substrate for in vitro incubation with intestinal slices
(cf. 168). Argentation TLC could then be used to isolate
triolein from the absorbed triglycerides. If this isolated
triolein had the same 3H :I4C ratio as the substrate triolein
then absorption of the triglyceride intact would be
unequivocally proved.
I n the general confusion, at the moment, as to which
lipids may be metabolically active and which merely
structural, it is beginning to appear likely that certain
individual molecular species of a given lipid class are
intimately concerned in metabolism while the bulk of
that lipid class present in the system may fulfill only a
passive structural role. Even a pure lipid such as lecithin
or phosphatidyl glycerol is, after all, a complex mixture of
compounds of differing physical properties and it is
only by studying the behavior of the individual molec-
ular species of such pure lipids that this type of problem
can be solved. The first steps along this road have already
been taken by van Deenen and his school in determining,
by argentation TLC and other methods, the individual
molecular species of phosphatidyl glycerol in photo-
synthetic tissues (138, 139) and of lecithin and other
phospholipids in animal tissues (136, 137) and in trying
to understand the implications of the proportions
and compositions of these in membrane structure and
function. This type of study will certainly be one of the
major avenues of research in lipid biochemistry in the
next few years and argentation chromatography will
have a major role to play in following it.
Manuscript received 7 7 May 7966.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
REFERENCES
Morris, L. J . Lab. Practice 13: 284, 1964.
Morris, L. J . I n Metabolism and Physiological Sign;fcance
of Lipids, edited by R. M. C. Dawson and D. N. Rhodes.
J. Wiley & Sons Ltd., London, 1964, p. 641.
Morris, L. J. In New Biochemical Separations, edited by
A. T. J ames and L. J . Morris. D. Van Nostrand Co.,
London, 1964, p. 295.
Den Boer, F. C. 2. Anal. Chem. 205: 308, 1964.
J urriens, G. Riv. Ital. delle Sostanze Grasse no vol: 116,
1965.
Mangold, H. K. J . Am. Oil Chemists Soc. 41: 762, 1964.
Mangold, H. K., H. H. 0. Schmid, and E. Stahl.
Methods Biochem. Analy. 12: 394, 1964.
Scholfield, C. R. In Fatty Acids, edited by K. Markley.
Interscience Publishers, New York, 1964, 2nd ed., p.
2283.
Radin, N. S. J. Am. Oil Chemists SOC. 42: 569, 1965.
Privett, 0. S., M. L. Blank, D. W. Codding, and E. C.
Nickell. J. Am. Oil Chmists SOC. 42: 381, 1965.
Pelick, N., T. L. Wilson, M. E. Miller, F. M. Angeloni,
and J . M. Stein. J. Am. Oil Chemists SOC. 42: 393, 1965.
Nichols, B. W., L. J. Morris, and A. T. J ames. Brit. Med.
Bull. 22: 137, 1966.
Morris, L. J ., and B. W. Nichols. In Chromatography,
edited by E. Heftmann. Reinhold Publishing Corp., New
York, 2nd ed., in press.
Winstein, S., and H. J . Lucas. J. Am. Chem. SOC. 60: 836,
1938.
Gil-Av, E., and J. Herling. J. Phys. Chem. 67: 1208,
1962.
MORRIS Argentation Chromatography 729

b
y

g
u
e
s
t
,

o
n

S
e
p
t
e
m
b
e
r

1
6
,

2
0
1
4
w
w
w
.
j
l
r
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

16. Muhs, M. A., and F. T. Weiss. J. Am. Chem. SOC. 84: 4697,
1962.
17. Lucas, H. J ., and coworkers. J. Am. Chem. SOC. (a) 65:
227, 230, 1943; (6) 74: 1333, 1338, 1952; (c) 76: 3931,
1954: ( d) 78: 1665, 1956; ( e) 79: 1306, 4339, 4341, 1957.
18. Andrews, L. J ., and R. M. Keefer. J. Am. Chem. Sac. 71:
3644, 1949; 72: 3113, 5034, 1950; 74: 640, 1952; 78:
2210, 1956.
19. Traynham, J. G., and coworkers. J. Am. Chem. SOC. 78:
4024, 1956; 81: 571, 1959.
20. Chatt, J . (chapt. 8), and G. Salomon(chapt. 9) in Cationic
Polymerisation and Related Complexes, edited by P. H.
Plesch. Academic Press, New York, 1953.
21. Coates, G. E. Organometallic Compounds. Methuen, London,
1960, 2nd ed., chapt. 6, p. 233.
22. Bennet, M. A. Chem. Rev. 62: 611, 1962.
23. Dewar, M. S. J. Bull. SOC. Chim. France 18: C79, 1951.
24. Taufen, H. J., M. J . Murray, and F. F. Cleveland. J.
25. Powell, D. B., and N. Sheppard. J . Chem. SOC. no vol:
26. Featherstone, W., and A. J. S. Sorrie. J. Chem. SOC. no
27. Nichols, P. L., J r. J. Am. Chem. SOC. 74: 1091, 1952.
28. Bradford, B. W., D. E. Harvey, and D. E. Chalkley.
29. Tenny, H. M. Anal. Chem. 30: 2, 1958.
30. Keulemans, A. I. M. Gas Chromatography. Reinhold Pub-
lishing Corp., New York, 1957, p. 205.
31. Bednas, M. E., and D. S. Russell. Can. J . Chem. 36: 1272,
1958.
32. Dutton, H. J ., C. R. Scholfield, and E. P. J ones. Chem.
Ind. (London) no vol: 1874, 1961.
33. Scholfield, C. R., E. P. Jones, R. 0. Butterfield, and H. J.
Dutton. Anal. Chem. 35: 386, 1963.
34. Scholfield, C. R., E. P. J ones, R. 0. Butterfield, and H.
J. Dutton. Anal. Chem. 35: 1588, 1963.
35. J ones, E. P., C. R. Scholfield, V. L. Davison, and H. J .
Dutton. J. Am. Oil Chemists SOC. 42: 727, 1965.
36. de Vries, B. Presented at VIth Congress, International
Society for Fat Research, London, April 1962; Chem.
Ind. (London) no vol: 1049, 1962.
37. Goering, H. L., W. D. Closson, and A. C. Olson. J. Am.
Chem. SOC. 83: 3507, 1961.
38. Wickberg, B. J. Org. Chem. 27: 4652, 1962.
39. Runeberg, J. Acta Chem. Scand. 14: 1985, 1960.
40. Enzell, C. Acta Chem. Scand. 15: 1303, 1961.
41. Barreto, H. S., and C. Enzell. Acta Chem. Scand. 15: 1313,
42. Daniels, P., and C. Enzell. Acta Chem. Scand. 16: 1530,1962.
43. Morris, L. J . Presented at VIth Congress, International
Society for Fat Research, London, April 1962; Chem.
Znd. (London) no vol: 1238, 1962.
44. Barrett, C. B., M. S. J. Dallas, and F. B. Padley. Chem.
Znc. (London) no vol: 1050, 1962.
45. Gunstone, F. D., F. B. Padley, and M. I. Qureshi. Chem.
Znd. (London) no vol: 483, 1964.
46. Gunstone, F. D., R. J. Hamilton, and M. I. Qureshi.
J. Chem. Sac. no vol: 319, 1965.
47. Subbaram, M. R., and Youngs, C. G. J. Am. Oil Chem-
ists Sac. 41: 445, 1964.
48. Wagner, H., J .-D. Goetschel, and P. Lesch. Helv. Chim.
Acta 46: 2986, 1963.
49. Anderson, R. L., and E. J. Hollenbach. J. Lipid Res. 6:
577, 1965.
Am. Chem. SOC. 63: 3500, 1941.
2519, 1960.
vol: 5235, 1964.
J. Inst. Petrol. 41: 80, 1955.
1961.
50. Willner, D. Chem. Ind. (London) no vol: 1839, 1965.
51. Wurster, C. F., J r., J. H. Copenhaver, J r., and P. R.
Schafer. J. Am. Oil Chemists SOC. 40: 513, 1963.
52. Emken, E. A., C. R. Scholfield, and H. J. Dutton. J. Am.
Oil Chemists SOC. 41: 388, 1964.
53. J ames, A. T., J . R. Ravenhill, and R. P. W. Scott. Chem.
Ind. (London) no vol: 746, 1964.
54. Urbach, G. J. Chromatog. 12: 196, 1963.
55. de J ong, K., K. Mostert, and D. Sloot. Rec. Trav. Chim.
56. Zinkel, D. F., and J . W. Rowe. J. Chromatog. 13: 74, 1964.
57. Badings, H. T., and J . G. Wassink. Neth. Milk Dairy J.
58. Morris, L. J. J. Lipid Res. 4: 357, 1963.
59. Cubero, J. M., and H. K. Mangold. Mimochem. J. 9: 227,
60. de Vries, B. J. Am. Oil Chemists SOC. 40: 184, 1963.
61. Stahl, E., and H. Vollman. Talanta 12: 525, 1965.
62. Klein, P. D., J . C. Knight, and P. A. Szczepanik. J. Am.
63. Stahl, E. Angew. Chem., Intern. Ed. Engl. 3: 784, 1964.
64. Morris, L. J., D. M. Wharry, and E. W. Hammond. J.
65. GrCen, K., and B. Samuelsson. J. Lipid Res. 5: 117, 1964.
66. Gupta, A. S., and S. Dev. J. Chromatog. 12: 189, 1963.
67. Barrett, C. B., M. S. J . Dallas, and F. B. Padley. J. Am.
68. Wood, R., and F. Snyder. J. Am. Oil Chemists SOC. 43:
69. Vereshchagin, A. G. J. Chromatog. 17: 382, 1965.
70. Paulose, M. M. J. Chromatog. 21: 141, 1966.
71. Privett, 0. S., and E. C. Nickell. J. Am. Oil Chemists SOC.
72. Dunn, E., and P. Robson. J. Chromatog. 17: 501, 1965.
73. Privett, 0. S., M. L. Blank, and 0. Romanus. J. Lipid
74. Bergelson, L. D., E. V. Dyatlovitskaya, and V. V. Voron-
75. Ruseva-Atanasova, N., and J . JanAk. J. Chromatog. 21:
76. Kuemmel, D. F. J. Am. Oil Chemists SOC. 41: 667, 1964.
77. Bhatty, M. K., and B. M. Craig. J. Am. Oil Chemists SOC.
78. Gunstone, F. D., and A. J. Sealy. J. Chem. Sac. no vol:
79. Craig, B. M., and M. K. Bhatty. J. Am. Oil Chemists Sac.
80. Powell, R. G., C. R. Smith, J r., and I. A. Wolff. J. Am.
81. Morris, L. J ., and M. 0. Marshall. Chem. Ind. (London)
82. Gunstone, F. D., and R. Subbarao. Chem. Znd. (London)
83. Gunstone, F. D., and A. J . Sealy. J. Chem. SOC. no vol:
84. Morris, L. J . J. Chem. SOC. no vol: 5779, 1963.
85. Powell, R. G., and C. R. Smith, J r. Biochemistry 5: 625,
86. Morris, L. J., and S. G. Welch, to be published.
87. Meijboon, P. W., and G. J urriens. J. Chromatog. 18: 424,
88. de Vries, B., and G. J urriens. Fette, Seifen, Anstrichmittel
89. Lees, A. M., and E. D. Korn. Biochim. Biophys. Acta 116:
82: 837, 1963.
17: 132, 1963.
1965.
Oil Chemists SOC. 43: 275, 1966.
Chromatog., in press.
Oil Chemists Sac. 40: 580, 1963.
53, 1966.
40: 189, 1963.
Res. 4: 260, 1963.
kova. J. Chromatog. 15: 191, 1964.
207, 1966.
41: 508, 1964.
4407, 1964.
41: 209, 1964.
Oil Chemists SOC. 42: 450A, 1965, abstract No. 35.
no vol: 460, 1966.
no vol: 461, 1966.
5772, 1963.
1966.
1965.
65: 725, 1963.
403, 1966.
730 JOURNAL OF LIPID RESEARCH VOLUME 7. 1966

b
y

g
u
e
s
t
,

o
n

S
e
p
t
e
m
b
e
r

1
6
,

2
0
1
4
w
w
w
.
j
l
r
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

90. Subbaram, M. R., and C. G. Youngs. J. Am. Oil
Chemists Soc. 41: 150, 1964.
91. Frankel, E. N., H. K. Peters, E. P. Jones, and H. J.
Dutton. J. Am. Oil Chi s t s SOC. 41: 186, 1964.
92. Frankel, E. N., E. P. Jones, V. L. Davison, E. Emken, and
H. J. Dutton. J. Am. Oil Chemists SOC. 42: 130, 1965.
93. J ardine, I., and F. J. McQuillin. J. Chem. SOC. no vol:
458, 1966C.
94. Mikolajczak, K. L., and M. 0. Bagby. J. Am. Oil Chem-
ists Soc. 42: 43, 1965.
95. Bagby, M. O., C. R. Smith, J r., and I. A. Wolff. Lipids
1: 263, 1966.
96. Morris, L. J ., and M. 0. Marshall. Chem. Znd. (London)
no vol: 1493, 1966.
97. Morris, L. J., M. 0. Marshall, and E. W. Hammond, to
bepublished.
98. Privett, 0. S., and E. C. Nickell. Lipids 1: 98, 1966.
99. de Jong, K., and H. Van der Wel. Nature 202: 553, 1964.
100. Morris, L. J ., M. 0. Marshall, and W. Kelly. Tetrahedron
101. Morris, L. J ., and D. M. Wharry. J. Chromatog. 20: 27,
102. Morris, L. J ., and D. M. Wharry. J. Chromatog., in press.
103. Morris, L. J ., R. T. Holman, and K. Fontell. J. Lipid
104. Tulloch, A. P. J. Am. Oil Chemists SOC. 41: 833, 1964.
105. Kishimoto, Y., and N. S. Radin. J. Lipid Res. 5: 94, 1964.
106. van Dorp, D. A., R. K. Beerthuis, D. H. Nugteren, and
H. Vonkeman. Biochim. Biophys. Acta 90: 204, 1964.
107. Bergstrom, S., H. Danielsson, and B. Samuelsson. Bio-
chim. Biophys. Acta 90: 207, 1964.
108. Hamberg, M., and B. Samuelsson. Biochim. Biophys.
Acta 106: 215, 1965.
109. Anggbrd, E., and B. Samuelsson. J. Biol. Chem. 240: 3518,
1965.
110. Nicholaides, N., and T. Ray. J. Am. Oil Chemists Soc. 42:
702, 1965.
111. Pohl, S., J . H. Law, and R. Ryhage. Biochim. Biophys.
Acta 70: 583, 1963.
112. Cornelius, J. A., and G. Shone. Chem. Znd. (London) no
vol: 1246, 1963.
113. Kircher, H. W. J. Am. Oil Chemists SOC. 42: 899, 1965.
114. Raju, P. K., and R. Reiser. Lipids 1: 10, 1966.
115. Haahti, E., T. Nikkari, and K. J uva. Acta Chem. Scand.
17: 538, 1963.
116. S&m, F., V. Wollrab, P. Jarolimek, and M. Streibl.
Chem. Znd. (London) no vol: 1833, 1964.
117. Wood, R., and F. Snyder. Lipids 1: 62, 1966.
118. Renkonen, O., 0. V. Renkonen, and E. L. Hirvisalo.
119. de Vries, B. J. Am. Oil Chemists SOC. 41: 403, 1964.
120. Dolev, A., and H. S. Olcott. J. Am. Oil Chemists SOC. 42:
121. de Vries, B., and G. J urriens. J. Chromatog. 14: 525,
122. J urriens, G., B. de Vries, and L. Schouten. J. Lipid Res.
123. J urriens, G., B. de Vries, and L. Schouten. J. Lipid Res.
124. Vioque, E., M. P. Maza, and M. Calderh. Grasas Aceites
125. Gunstone, F. D., and F. B. Padley, J. Am. Oil Chemists
126. Gunstone, F. D., and M. I. Qureshi. J. Am. Oil Chemists
Letters no vol: 4249, 1966.
1965.
Res. 1: 412, 1960.
Acta Chem. Scand. 17: 1465, 1963.
624, 1965.
1964.
5: 267, 1964.
5: 366, 1964.
15: 173, 1964.
Soc. 42: 957, 1965.
SOC. 42: 961, 1965.
127. Kaufmann, H. P., and H. Wessels. Fette, Seifen, Anstrich-
128. Blank, M. L., and 0. S. Privett. J. Dairy Science 47: 481,
129. Litchfield, C., M. Farquhar, and R. Reiser. J. Am. Oil
130. J urriens, G., and A. C. J. Kroesen. J. Am. Oil Chemists
131. Blank, M. L., B. Verdino, and 0. S. Privett. J. Am. Oil
132. Blank, M. L., and 0. S. Privett. Lipids 1: 27, 1966.
133. Morris, L. J . Biochem. Biophys. Res. Commun. 20: 340,
1965.
134. Kaufmann, H. P., H. Wessels, and C. Bondopadhyaya.
Fette, Seifen, Anstrichmittel 65: 543, 1963.
135. Arvidson, G. A. E. J. Lipid Res. 6: 574, 1965.
136. van Golde, L. M. G., R. F. A. Zwaal, and L. L. M. van
Deenen. Koninkl. Ned. Acad. Wetemchap., Proc. Series B,
68: 255, 1965.
137. van Deenen, L. L. M., L. M. G. van Golde, and R. A.
Demel. Biochem. J. 98: 17P, 1966.
138. Haverkate, F. Doctoral Thesis, Phosphatidyl glycerol from
photosynthetic systems, University of Utrecht, Holland, 1965.
139. Haverkate, F., and L. L. M. van Deenen. Biochim. Bio-
phys. Acta 106: 78, 1965.
140. Renkonen, 0. Acta Chem. Scand. 18: 271, 1964.
141. Renkonen, 0. J. Am. Oil Chemists SOC. 42: 298, 1965.
142. Renkonen, 0. Lipids 1: 160, 1966.
143. Avigan, J., D. S. Goodman, and D. Steinberg, J. Lipid
144. Truswell, A. S., and W. D. Mitchell. J. Lipid Res. 6: 438,
145. Claude, J . R. J. Chromatog. 17: 596, 1965.
146. Kkan, R., and M. Cudzinovski. J. Chromatog. 18: 422,
147. Lees, T. M., M. J . Lynch, and F. R. Mosher. J . Chroma-
148. Copius-Peereboom, J . W. Z. Anal. Chem. 205: 325, 1964.
149. Copius-Peereboom, J . W., and H. W. Beekes. J. Cho-
150. Vlachov, R., and M. Holub. Coll. Czech. Chem. Commun.,
151. Zabza, A., M. Roman&, and V. Herout. Coll. Czech.
152. Herout, V., A. Banassek, and M. Romanuk. Coll. Czech.
153. Norin, T., and L. Westfelt. Acta Chem. Scand. 17: 1828,
154. Nano, G. M., and A. Martelli. J. Chromatog. 21: 349,
155. Ikan, R. J. Chromatog. 17: 591, 1965.
156. Schreiber, K., 0. Aurich, and G. Osske. J . Chromatog. 12:
157. Ercoli, A., R. Vitali, and R. Gardi. Steroids 3: 479,1964.
158. Harris, R. V., L. J. Morns, and A. T. James, to bepub-
159. Davidoff, F., and E. D. Korn. J. Biol. Chem. 239: 2496,
160. Struijk, C. B., and R. K. Beerthuis. Biochim. Biophys. Acta
161. Schroepfer, G. J., and K. Bloch. J. Biol. Chem. 240: 54,
162. Harris, R. V., P. Harris, and A. T. James. Biochim. Bio-
mittel 66: 81, 1964.
1964.
Chemists SOC. 41: 588, 1964.
SOC. 42: 9, 1965.
Chemists SOC. 42: 87, 1965.
Res. 4: 100, 1963.
1965.
1965.
tog. 18: 595, 1965.
matog. 17: 99, 1965.
in press.
Chem. Commun. 31: 3012, 1966.
Chem. Commun., in press.
1963.
1966.
63, 1963.
lished.
1964.
116: 12, 1966.
1965.
phys. Acta 106: 465, 1965.
MORRIS Argentatwn Chromdfography 731

b
y

g
u
e
s
t
,

o
n

S
e
p
t
e
m
b
e
r

1
6
,

2
0
1
4
w
w
w
.
j
l
r
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

163. Nichols, B. W., P. Harris, and A. T. J ames. Biochem. 166. Ward, P. F. V., T. W. Scott, and R. M. C. Dawson.
164. Sgoutas, D. S., and F. A. Kummerow. J. Chromatog. 16: 167. Wilde, P. F., and R. M. C. Dawson. Biochem. J. 98: 469,
165. Cvetanovic, R. J., F. J . Duncan, and W. E. Falconer. 168. Feldman, E. B., and B. Borgstrom. Lipids 1: 128,
Biophys. Res. Commun. 21: 473, 1965.
448, 1964. 1966.
Can. J. Chem. 41: 2095, 1963. 1966.
Biochem. J. 92: 60, 1964.
732 JOURNAL OF LIPID RESEARCH VOLUME 7, 1966

b
y

g
u
e
s
t
,

o
n

S
e
p
t
e
m
b
e
r

1
6
,

2
0
1
4
w
w
w
.
j
l
r
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m